At Craftworkz, we always try to keep up with all the latest technologies. Furthermore, we seem to be excited about new ways for humans to interact with a computer system. After all, we have a lot of experience with chatbot technology and voice assistants, together with our colleagues from Oswald. This time, we wanted to examine where the world stands with brain-computer interfaces or BCI.
A BCI is used to control computer system with your mind, which kind of sounds like the plot of a lame sci-fi movie from the late 1980s, but because of new start-ups investing a lot in the hardware, software engineers like us can start experimenting with these new tech devices. For our experiments, we chose the Emotiv Epoc+
Introduction to EEG
EEG stands for Electroencephalography and it is a method to measured electrical activity generated by the synchronized activity of thousands of neurons. This is displayed in volts. As the voltage fluctuations measured at the electrodes are very small, the recorded is digitized and sent to an amplifier. The amplified data can then be displayed as a sequence of voltages values. At first this method was only used in hospitals for research, but lately some companies are creating EEG headset for consumers. Some examples of these companies are Emotiv, Muse, OpenBCI, Neurosky,… This post is based on the Epoc+ headset from Emotiv.
The Emotiv Epoc+ is a mobile EEG headset that consists of 14 channels and 2 references. These channels are also known as sensors and are made of saline-soaked felt pads that are placed at a specific point along the scalp. The EEG signals have a sampling rate of 2048 Hz with 14 bits of resolution. The headset is supported by Windows 7+ platforms, MacOS, iOS 9+ and Android 4 and above.
Emotiv offers three kinds of Detection algorithms, all of them are built on extensive scientific studies to develop accurate machine learning algorithms to classify and grade the intensity of different conditions.
The headset currently measures 6 different cognitive states in real time:
- Excitement (arousal)
- Interest (valence)
- Stress (frustration)
These detections are based on rigorous experimental studies from Emotiv.
This is a system where the user can train direct mental commands. The user trains the system to recognize thought patterns related to different desired outcomes. Such as moving an object or making it disappear.
Muscle artifacts which commonly get rejected in laboratory EEG studies are diverted and classified to map the activation in different muscle group and eye movement events.
Some prototypes we've built with it
The first demo I ever built with the headset was the game space invaders that you could control with facial commands. You would raise your eyebrows to move the spaceship to the left, clench your teeth to go right and blink with your left eye to shoot.
IOT with your brain
The second prototype was to control the physical world with your mind. I Created a simple IOT demo where a user could train a mental command and use that command to turn on a led-light that was connected to a raspberry pie. To make this all possible we had to run a web server on the raspberry that would get listen to the request that was made from the headset. After it would get a specific request it would change the behavior of the led-light.
Use cases with BCI
There are endless use cases in various sectors where an EEG headset would be useful . For example marketeers are eager to start doing research on the cognitive states of their customers when in contact with their brand. In the healthcare sector there are many possibilities from brain controlled wheelchairs for disabled patients. To preventive scans for brain diseases like dementia.The gaming industry is always looking to improve their controllers. So it won't be a surprise if they are doing research on BCI's
Working with new technologies like this is exciting. Personally, I never heard about this EEG headset and — to be honest — at first, it looked very daunting to write a program that you could control with your mind. However, after some weeks of research and building prototypes, It turned out to be way easier than I would have imagined. At Craftworkz, we see a great future for BCI’s in many different sectors, from healthcare to gaming, to marketing… and we will keep our team up to date with this technology. However, at this point, we think the technology still needs some time to mature (sometimes it misreads your thought patterns, especially if you’re training multiple mental commands). Even though the tech still needs some time to evolve to become a part of our daily lives, we’re already eager to start implementing some exciting business cases